当前位置: X-MOL 学术Int. J. Pattern Recognit. Artif. Intell. › 论文详情
Our official English website, www.x-mol.net, welcomes your feedback! (Note: you will need to create a separate account there.)
Leakage Signal Analysis of Urban Gas Pipeline Based on Improved Variational Mode Decomposition
International Journal of Pattern Recognition and Artificial Intelligence ( IF 1.5 ) Pub Date : 2020-01-31 , DOI: 10.1142/s0218001420580185
Yongmei Hao 1 , Zhanghao Du 1 , Zhixiang Xing 1 , Xiaohu Mao 2
Affiliation  

Aiming at problems of multipoint leakage source detection and low positioning accuracy in urban gas pipelines, a multipoint leak location method base on improved variational mode decomposition (VMD) was proposed. By improving the VMD decomposition of the original leakage signal, the parameters of the VMD were optimized to reduce the influence of noise and weak correlation signals on the leak location. Then the multi-point leakage location model of pipeline was established, and the sensitive modal component Intrinsic mode function (IMF) with the most leakage information was selected by multiscale entropy. According to the characteristics of the blind source separation method, the relevant time delays of the simultaneous leakage of multiple points on the pipeline and the frequency of the signal are extracted. Finally, The location of the leak source is determined according to the principle of cross-correlation. The experimental results show that compared with the direct cross-correlation method and the VMD-based method, the proposed multipoint leak diagnosis method has less error, the minimum relative error is 1.61%, and the positioning accuracy is higher.

中文翻译:

基于改进变分模态分解的城市燃气管道泄漏信号分析

针对城市燃气管道多点泄漏源检测和定位精度不高的问题,提出了一种基于改进变分模态分解(VMD)的多点泄漏定位方法。通过改进对原始泄漏信号的VMD分解,优化VMD参数,减少噪声和弱相关信号对泄漏位置的影响。然后建立管道多点泄漏定位模型,通过多尺度熵选择泄漏信息最多的敏感模态分量固有模态函数(IMF)。根据盲源分离法的特点,提取管道上多点同时泄漏的相关时延和信号频率。最后,根据互相关原理确定泄漏源的位置。实验结果表明,与直接互相关法和基于VMD的方法相比,所提出的多点泄漏诊断方法误差更小,最小相对误差为1.61%,定位精度更高。
更新日期:2020-01-31
down
wechat
bug